The purification of blood and separation of fluids using dialysis can be advantageously used in many medical applications, particularly conditions where renal function has significantly declined. Dialysis removes wastes from blood through a semipermeable membrane by diffusive or convective processes. There are two principal dialysis methods used to support patients requiring renal replacement therapy: hemodialysis and peritoneal dialysis
Hemodialysis, involves the removal of solutes and fluids (such as urea, creatinine and uric acid) from the blood through a dialysis membrane by diffusion into a dialysate. The dialysis membrane is a semipermeable membrane which is typically made of cellulose. Blood solutes containing the waste permeate through the membrane and into a dialysis solution or dialysate formulated to control solute net movement through the membrane.
In the chronic hemodialysis setting, processes which have been developed and are commonly used provide bicarbonate dialysis using a highly sophisticated machine which can be monitored by a team. Dialysis provided in the intensive care setting for patients with an acute loss of kidney function has traditionally been provided with a chronic hemodialysis machine, brought into the unit and operated by one dialysis nurse per patient, in addition to the patient's intensive care nurse.
Hemodialysis can be either continuous or intermittent. Intermittent hemodialysis involves short intensive periods of treatment on alternate days, while continuous hemodialysis involves continuous fluid removal and continuous blood purification.
Peritoneal dialysis, which also uses reverse osmosis principles, is another procedure which is used to remove waste products from a patient. This type of dialysis uses the peritoneal lining of the patient's abdomen as a semipermeable membrane to filter blood. In peritoneal dialysis, peritoneal dialysate is infused into the patient's peritoneum through a catheter. Fluid and waste removal is achieved by an osmotic gradient from the blood to the dialysate, generated by a high glucose concentration, permitting water to flow out from the blood. Fluids and waste products pass from the many blood vessels and capillaries in the peritoneal membrane into the dialysate, and after a sufficient period of time the dialysate containing the fluids and waste products is drained from the abdomen.
Chronic peritoneal dialysis is most often performed by the patient alone in their home after suitable training in a dialysis center, and Involves the use of individually applied bags of ready made, sterile, lactate buffered dialysis solution.
Due to resource limitations dialysis often must be condensed into a period of hours and may be limited to less than daily treatments leading to large fluctuations in levels of the substances removed front the patient. These fluctuations may adversely affect patient outcomes. A dialysis therapy which comes closest to normal kidney function, by operating continuously may improve patient outcomes and shorten intensive care stays, This has led to the adoption of continuous modalities of renal replacement therapy (CRRT) in the intensive care setting.
Continuous renal replacement therapy (CRRT) is dialysis continued 24 hours a day. Unlike chronic haemodialysis there are no standardized equipment or processes for CRRT. To simplify the equipment necessary, CRRT does not use dialysate from concentrate, but uses pre-made dialysate, usually peritoneal dialysis solution. This solution is sterile and is buffered by lactate. The dialysis solution to which blood is exposed through this membrane should have the same electrolyte composition of normal serum or it may induce fatal electrolyte abnormalities. Its use with dialysis filters requires at a minimum the absence of pyrogens. If the solution is to be given intraperitoneally or intravenously it must be sterile and pyrogen free.
The electrolyte composition of all dialysis solutions may vary but in a narrow range. The major cationic electrolyte component is sodium, usually at the concentration it is found in serum 140 (mmol/L, mEq/L). Other cations include calcium (2.5 mmol/L, 5.0 mEq/L, 10 mg/dl) and magnesium (0.75 mmol/L, 1.5 mEq/L,.times.mg/dl). The major anion is chloride whose concentration is determined by the net of the cationic charge constituents less the anionic buffer. The dialysis solutions used in all forms of dialysis contain buffers in an attempt to correct metabolic acidosis. Common buffers used include bicarbonate, lactate and acetate buffers.
Bicarbonate buffer is a preferred buffer for dialysis since bicarbonate is the physiological buffer of the body. However, pre-made mixtures of bicarbonate buffered solutions are difficult to sterilize and store because released carbonate will precipitate with calcium if present. Attempts have been made to stabilize calcium, for example with glycylglycine (U.S. Pat. No. 5,211,643 to Reinhardt et al). Continuous dialysis against an agent such as glycylglycine produces levels in the blood close to those present in the dialysate. The effect of long term exposure to stabilizing agents such as glycylglycine is unknown (Yatzidis et al. Nephron., 64:27-31, 1993).
Furthermore, sugars in a dialysis solution will caramelize during heat sterilization and prolonged exposure if kept at neutral or higher pH (7.4). Therefore sugar containing dialysis solution is kept at low pH. For example, pH 5.4 for most peritoneal dialysis solutions. The low pH is believed to be the source of pain patients suffer after instillation of a fresh bag of peritoneal dialysis solution. Low pH solutions are known to reduce the effectiveness of peritoneal immnunologic defences. The safety of using low pH solutions for dialysis or hemofiltration during CRRT has not been studied.
Also, during preparation and storage of a bicarbonate buffered solution, CO.sub.2 is released from the solution, changing the bicarbonate concentration and pH of the solution. It is therefore necessary for bicarbonate containing solutions to be stored in glass or CO.sub.2 impermeable plastic containers. The following solutions have been proposed to control the CO.sub.2 content of the bicarbonate solution for peritoneal dialysis: storage in a powder form until use; use of an impermeable barrier between calcium containing and bicarbonate containing portions; and addition of buffers such as histidine or glycylglycine (H. Yatzidis, Nephron 64:27-31, 1993).
Dialysis care has become process driven to maximize the quality of the dialysis and to minimize costs. Hemodialysis machines have been developed which can prepare dialysis solution online from a single concentrate and clean water provided from a central reverse osmosis system. To get around the stability problems associated with calcium and bicarbonate, acetate was substituted for bicarbonate. Acetate hemodialysis was carried out until evidence showed the deleterious effects of acetate on dialysis patients, particularly with the use of the newer more biocompatible dialysis membranes (F. H. Leenen, Artificial Organs 8:411-417, November 1994).
Dual proportioning dialysis machines have been developed and employed at great expense to provide bicarbonate dialysis. These machines solve the calcium bicarbonate instability problem by keeping the bicarbonate and acid concentrates separate until the time of dialysis. Although micro precipitation may occur immediately after mixture, clinically this is not a concern even over a 72 hour period (Leblanc et al, 1995). However, because of this precipitation bicarbonate dialysis machines must have acid rinses on a regular basis. Separate batches of concentrates have been used using split bags which contain calcium and magnesium on the one hand, and the bicarbonate on the other hand to prevent precipitation (U.S. Pat. No. 4,630,727 to Feriani et al).
A method was been developed to allow an older single proportioning chronic dialysis machine to produce bicarbonate dialysis from concentrate using calcium free bicarbonate concentrate adding the calcium back into the blood by an infusion pump. This method for chronic dialysis was reported by Kaye et al, but was not adopted outside of Kaye's unit in Montreal. (M. Kaye et al., Clinical Nephrology 31:132-138 1989; M. Kaye and D. Fisher, Clinical Nephrology 34:84-87, 1990; and M. Kaye, Clinical Nephrology 40:221-224, 1993). Calcium is infused distal to the dialyzer into the drip chamber using an infusion pump and is a component of the dialysate. In Kaye's studies, the patient's are not critically ill and his system is set up for chronic hemodialysis, not for acute hemodialysis. The concentrate used by Kaye is not sterile. Furthermore, Kaye's system is used for intermittent, but not for continuous dialysis.
Acute renal failure in critically ill patients, which is generally accompanied by metabolic derangements and high overall mortality, poses significant challenges for renal replacement therapy. Acute intermittent hemodialysis has been the conventional therapy. Bicarbonate dialysate which is typically used in acute intermittent hemodialysis is not sterile but only clean.
Problems with the rapid removal of fluid and changes in electrolytes which occur during high efficiency short term intermittent hemodialysis have led to the development and use of continuous renal replacement therapies (CRRT) for critically ill patients (P. Y. W. Tam et al., Clinical Nephrology 30:79-85, 1988 and E. F. H. Van Bommel et al, Am. J. Nephrol. 15:192-200, 1995). Solute and volume removal are slow and continuous during CRRT eliminating the large shifts occurring between body compartments during intermittent hemodialysis, which may lead to hypotension and interfere with renal recovery (E. F. H. Van Bommel, Nephrol. Dial. Transplant. 1995 Editorial Comments, p. 311). CRRT techniques include peritoneal dialysis, continuous arterio-venous and veno-venous ultrafiltration, hemofiltration, hemodialysis and heimodiafiltration. Traditionally CRRT has used peritoneal dialysis solution as the dialysate and infusate.
Lactate containing peritoneal dialysis solution has been used in CRRT dialysate with some success (Baxter and Gambro solutions). Lactate is stable with calcium and is stable at low pH (5.4). Lactate is metatbolised by the intact functioning liver into bicarbonate, the body's natural buffer. However, lactate infusions are known to induce panic in susceptible individuals and may alter metabolism to favour catabolism over anabolism (R. L. Veech et al.). Its safety in CRRT dialysis has not been tested. However, its use as a buffer in peritoneal dialysis solution is universal and appears to be tolerated, except for abdominal pain and possible immunologic effects; there is mounting evidence that exposure to large amounts of lactate, particularly in the racemic form, may not be benign. Lactate included in these solutions is of the racemic form.
In intensive care patients, such as patients who have developed hypotension and lactic acidosis, lactate from the dialysis solution may not be metabolized to bicarbonate because of liver dysfunction, and when the dialysate lacks bicarbonate, acidosis may be worsened due to bicarbonate removal during dialysis. (A. Davenport et al., Nephron 1991:59:461-465, 1991 and M. Leblanc et al., Am. J. Kid. Dis. 26:910-917, 1995). For acute hemodialysis in the intensive care unit CRRT typically uses lactate based sterile solutions as dialysate and infusate (peritoneal dialysis solution). Research into methods to provide bicarbonate dialysate have been ongoing, Recently, a method was reported for providing non-sterile calcium bicarbonate dialysate for patients in the intensive care undergoing CRRT (M. Leblanc, AJKD 26(6):910-917, 1995). Non-sterile bicarbonate dialysis solutions can be produced in the chronic hemodialysis unit and carried to the intensive care unit. These methods are labour intensive, unregulated, non sterile, not pyrogen free, expensive and may lack sufficient quality control. Unlike chronic hemo- or peritoneal dialysis, which are process driven and carried out in a uniform, cost effective quality controlled manner, CRRT is carried out in many different modalities specific to each intensive care unit.
It is important to use a sterile dialysis solution in CRRT in order to avoid pyrogenic reactions caused by bacteria and endotoxin contamination of the dialysate solution. It is also important to have a solution which is readily available for use. While sterile lactate or acetate-based dialysis solutions may be used in CRRT they suffer from the disadvantages discussed above. It has been suggested that bicarbonate dialysate may be preferable to lactate or acetate-based solutions (M. Leblanc et al., Am. J. Kid. Dis, 26:910-917, 1995). However, it has not been possible to provide a sterile and readily available bicarbonate solution for CRRT due to the problems discussed above with bicarbonate solutions.